1,5,5-TRIMETKYLBICYCLO[2.1.1]HEXANE-6-CARBOXYLIC ACID
March 20, 1962
[CONTRIBUTION FROM
THE
977
DEPARTMENT OF CHEMISTRY, CORNELL UNIVERSITY, ITHACA, N. Y.]
Highly Strained Bicyclic Systems. V1.l Reactions of 1,5,5-Trimethylbicyclo[2.1.1]hexane-6-carboxylic Acid and the Synthesis of Bicyclo [2.1.1 ]hexyl Analogs of Borneol and Isoborneo12 B Y JERROLD ~ ~ E I N W A L D ARTHUR , LEWISAND PAUL G. Gasshf;iru RECEIVED AUGUST2, 1961 This work was undertaken t o elucidate the stereochemistry of Horner’s l,~,5-trimeth~-lbicyclo[2.l.l]l~exane-6-carbox~-l~c acid (11), and to develop syntheses of the highly strained, cyclobutanoid analogs (111, I\:) of borneol and isoborneol (V, VI). The following results or conclusions were reached: (1) Isocyclenone (VII) and camphoric anhydride (YIII) are neutral byproducts in the photochemical ring contraction of diazocamphor ( I )to 11. ( 2 ) This photolysis yields I I a ( e m ) as the chief reaction product, (3) I I a is epimerized t o I I b ( e n d o ) during hydrolysis of its methyl ester. ( 4 ) Both I I a and I I b can be readily degraded t o the common olefin X. (5) Both I I a and I I b are isomeriz:d by sulfuric acid t o dihydro-p-campholenolactone (XIII, 5Oy0)and a rearranged acid ( 2 0 c i ) ,previously considered a n epimer” of 11. (6) This rearranged acid may be assigned structure XXIIIa on the basis of chemical and physical evidence and independent synthesis. ( 7 ) Methyllithium converts I I b into the methyl ketone XXVI, which yields the endo-alcohol I11 on trifluoroperacetic acid oxidation followed by lithium aluminum hydride reduction. ( 8 ) Sarett oxidation of 111 yields the cyclobutanone XXVII, which is reduced t o the exo-alcohol IV by lithium aluminum hydride.
wish to describe the results of a more extensive Introduction Horner and Spietschka’s discovery that photol- study of the chemistry of Horner’s acid, undertaken ysis of diazocamphor (I) in aqueous medium gives with the aims of clarifying its stereochemistry and a 1,5,5-trimethylbicyclo [2.1.l]hexane-G-carboxylicof developing paths to I11 and IV. acid (11)3 provided the first ready entry into the Discussion bicyclo[2.l.l]hexane series. One of our objectives The photocheniical ring contraction of I was in studying the chemistry of these bridged com- carried out as described previously to give I1 (m.p. pounds was to evaluate the effects of skeletal strain 1OO-10lo, [ a ] ~ 13,s’) in about 607c yield. 111 and unusual geometry on their pr0perties.l To- addition, about 20% of a neutral fraction was ward this end, a synthesis of 1,5,5-trimethylbicyclo- isolated. Vapor phase chromatography of this [2.l.l]hexan-endo-G-o1 (111) and 1,5,5-trimethyl- fraction showed the presence of a t least two bicyclo [2.l.l]hexan-exo-6-01(IV), the cyclobutanoid major constituents, which were separated by fracanalogs of borneol (V) and isoborneol (VI), ap- tional sublimation and crystallization The more peared especially i n t e r e ~ t i n g . ~The resistance of volatile component proved to be isocyclenone (VII), Horner’s acid (11) to a-bromination and to Schmidt formed by a typical carbene insertion reaction. k’ The less volatile component was found to be camphoric anhydride (1711) , apparently resulting from reaction of I or the corresponding carbene with dissolved oxygen.
+
I11
v
01I
t\’
H
VI
degradation has been reported; both reactions gave an isomeric acid, regarded as the C6-epirner of 11, in place of the anticipated p r o d ~ c t s . 1 ~X;e (1) For Parts I V and V of this series see 5. Rleinwald and A. Lewis, J . A m . Chem. SOL., 83, 2769 (19611, and J. Meinwald, Rec. Chem. P i o g . , 22, 39 (1961). (2) Presented before the Division of Organic Chemistry of the American Chemical Society at the 139th h-ational Meeting in St. Louis, hlo., March 21-30, 1961 (abstract page 3Y-0). See also J. Meinwald, A. Lewis and P. G. Gassman, J . A m . Chcm. Soc., 82, 2549 (1960), for a preliminary account of part of this work. (3) L. Horner and E. Spietschka, Chem. B P I . .88, , 934 (195;). ( 4 ) Independent work aimed a t a study of the analogs of e m - and -norborne01 has been carried out by K. B. Wiberg, B. R. Lowry T. H. Colby, J . A m . Chcm. SOC.,8 3 , 3998 (1961).
In coilsidering the problem of the Stereochemistry of Horner’s acid, we anticipated from work in a related series that the intermediate ketene IX would suffer hydration from the less hindered side giving predominantly the unstable epimer of II.5 To test this point, the corresponding methyl ester was prepared and heated for a prolonged period with sodiutn methoxide in Xlthough the product recovered from this treatment was only unchanged ester, saponification using excess base in aqueous ethanol gave a new acid (m.p. 43.0-43.5O, [ a ] D +68.7O) in excellent yield. Apparently under these conditions the desired epimerization proceeded rapidly, competing favorably with ester h y d r ~ l y s i s . ~ The properties of our new acid were markedly different from those of Horner’s “epimer” (m.p. ( 5 ) J. Meinwald and P. G. Gassman, ibid., 82, 2857 (1960). (6) For a recent example of the use of this type of procedure, see G. W, Eigenmann and R. T. Arnold, ibid., 81, 3440 (1959). (7) A closely related pattern of behavior was described by J. Meinwald and P. G. Gassman, ibid., 82, 5445 (1960).
JERROLD MEINWALD, ARTHURLEWISAND PAULG. GASSMAN
978
80-8lo, [CZ]D -26.9'), requiring that one of these two acids be a rearrangement product. The treatment with concentrated sulfuric acid (or with hot bromine and phosphorus) seemed more suspect, and evidence to show that the original 100' acid and its new 43' isomer share a common carbon skeleton was successfully obtained as outlined below. 3,LiAlHb 4
H . 0 9
5,O.
Vol. 84
'OH
!ia or 11
steps
qCH* s
Each acid was converted into its acid chloride, dimethylamide, t-amine and amine oxide. Cope pyrolysis8 of both amine oxides proceeded very readily, and both gave rise to the highly volatile, crystalline olefin X . Structure X, once assigned to ~ a m p h e n e ,was ~ supported by analytical data and infrared evidence (maxima a t 3.29, 5.96 and 11.56 p, which vanish on treatment with bromine). I n deciding which of these two acids should be assigned the exo configuration IIa and which the endo configuration IIb, use was made of invaluable tables published by Wilcox.lo From these, the distances between the center of the carboxyl carbon and the edge of the nearest neighbor from which an unfavorable non-bonded interaction would be expected are found to be 0.80 A. for I I a and 1.12 A. for IIb. Thus, IIa appears significantly destabilized by non-bonded repulsions. Since Horner's 100' acid is efficiently isomerized to its 43' epimer, this isomerization can be confidently formulated as I I a + IIb, and the stereochemical problem is solved.
11 with an authentic sample of XI11 revealed their identity; confirmation was provided by the fact that both samples gave the same diol (XIV) upon lithium aluminum hydride reduction.'* I
XI11
v
I
XVI
Driving force for this rearrangement is probably provided by the relief of strain accompanying rupture of the folded cyclobutane ring, although we have shown in other work that this cleavage is not general for bicyclo [a.1.11hexanes.l Turning now to the acid formed along with XIII, possible deprotonation of intermediates X I or XI1 leads to consideration of composite formulas XV and XVI. The failure of the acid to react with tetranitromethane, bromine, potassium permanganate and ozone, however, immediately excludes these possibilities and requires a bicyclic structure. The close similarity between the isomerization conditions and those of the cinenic acid rearrangement (XVII + XVIII) , known to involve decarbonylation, rearrangement and recarbonylation,l 3 suggested analogous paths lending to two bicyclo[3.1.0]hexanes (XIX, XX) or two bicyclo [2.2.0]hexanes (XXI, XXII). Although n.1n.r. invcstigatioii of the SOo acid failed to yield Before turning to the work on degradative reactions, i t is apprppriate to consider the previously any helpful information, end-absorption in the misinterpreted rearrangement of I I a to an isomeric ultraviolet spectrum of the acid suggested conjuacid. Treatment of IIa (or IIb) with sulfuric acid gation between the carboxyl group and a cyclowas found to give 20y0 of the reported acid (xn.p. propane ring. To bring the suspected chromophore 80-81', [Q]D -26.9"), along with 50% of an un- into a less ambiguous spectral region, the acid was converted to the corresponding methyl ketone noticed neutral isomer (m.p. 38-37', [ a ] D 4-3.9'). A strong 5.64 12 infrared maximum suggested form- with metliyllithium. This ketone showed clear 206 ulation of the neutral product as a y-lactone, and spectral indications (Amax 5.90 / . L ~ ~ ~ ; m p (E 5,700), 278 mp ( E 67)14b) of cyclopropyl the reaction sequence outlined below led to the hypothesis that it was dihydro-/3-campholenolac- carbonyl conjugation. Its 2,4-dinitrophenylhytone (XIII).I1 Direct comparison of this lactone drazone had its long wave length maximum a t 370 mp, in accordance with expectations for this chroin(8) For a recent reference to the use of this technique, see A. C. ophore.l5 The only readily derivable structure Cope, E. Ciganek, C. F. Howell and E. E. Schweizer, J . Am. Chem. Soc., 82,4663 (1960). (12) We are grateful to Dr. Overton for supplying authentic samples (9) 0. Aschan," Die Constitution des Ramphers und Seiner Wichtigsten Derivate," Friedrich Vieweg und Sohn, Braunschweig, 1903, offers a very interesting account of the status of this area of terpene chemistry at the turn of the century. (IO) C. F. Wilcox, Jr., J. A m . Chem. Soc., 82, 414 (1960). (11) J. D. Connolly and K. H. Overton, Proc. Chem. Soc., 188 (1959)
of XI11 and XIV. (13) J. Meinwald, H.C. Hwang, A. P. Wolf and D. Christman, J . Am. Chem. Soc., 82, 483 (1960). (14) (a) G.W. Cannon, A. A. Santilli and P. Shenian, ibid., 81, 1660 (1959). (b) For a recent example and lending reference, see D. H. R . Barton and P. T. Gilham, J . Chem. Soc., 4596 (1960) (15) I. D.Roberts and C. Green, ibid., 68, 214 (1946).
1,s)5-TRIMETHYLBICYCLO[2.1,1]HEXANE-6-CARBOXYLICACID
March 20, 1962
979
I
1.679,~= Cyclopropyl -C-H I
I
XVII
COzH XVIII
3
COzH
xx
--f
I
h-+ 1.600 XXI
I *
+&
fitting these requirements is XXIII, arising from ion XI1 by C-alkylation (rather than the 0alkylation process which yields XIII) . Further treatment oi‘ XXIII with sulfuric acid converts
XI1
H XXIII
I
1
1.700
1.800
I
l.9OOp.
Fig. 1.-Near infrared spectra of XXIII (dotted line; 250 mg. in 34 ml. of CC&) and XXIV (so!id line; 208 mg. in 34 ml. of CCI4) as recorded on a Cary model 14 spectrophotometer.
(Fig. l), providing strong evidence for the presence of two cyclopropyl hydrogens in XXIII. It is noteworthy that in this instance, near-infrared spectroscopy serves where n.m.r. fails, since a tertiary cyclopropyl proton does not appear at the unusually high field characteristic of the cyclopropyl methylene protons. lS t N2CHC02C2H,
COJI
€1 i t into XIII. a process which is readily interpreted XXIV as an acid-catalyzed opening of the cyclopropane Finally, definitive chemical evidence for the strucring, followed by normal lactonization. Additional physical evidence supporting struc- ture of Horner’s rearrangement product was proture X X I I I was derived from examination of its vided by an independent synthesis of ( *J-XXIII, near-infrared spectrum. a technique which has carried out by employing ( j=)-1,3,S-trimethylcyclomuch to contribute to structural problems,I6 pentene in place of 1-niethylcyclopentene in the a!though i t has not yet becon-e one of the standard diazoacetic ester reaction used earlier to obtain tools in structural investigations. I t is known the model XXIV.lg The reported melting point that cyclopropyl methylene groups show character- for Horner’s racemic product (125’) agrees with ~ . ~ ’that found for (*)-XXIII (123-124’), and the istic first overtones in the 1.64 1.1 r e g i ~ n , ~but data for tertiary cyclopropyl hydrogens do not solution inirared spectra of the synthetic acid seem to be available. We find that XXIIL shows and methyl ester are superimposable on those oba characteristic peak a t 1.679 1.1 a t significantly tained from the levorotatory rearrangement prodshorter wave length than any saturated CH ab- uct. Although the question of the stereochemistry sorptions other than those of three-membered rings. of X X I I I has not been studied explicity, the conX variety of bicyclo[2.l.l]hexanes do not show ditions involved in its formation from I1 and in absorption in this region, so that this band cannot its independent synthesis permit the assignment be attributed to the CH of a strained cyclobutyl of the less crowded 6-exo configuration XXIIIa ring. Since the intensity of this band is related with some confidence. to the number of cyclopropyl hpdrogens,I6 the model compound XXIV was synthesized as shown. Equimolar solutions of X X I I I and XXIV revealed well-resolved peaks a t 1.679 p of equal intensity (16) An intriguing account of the present status of near-infrared spectroscopy is presented by R. F. Goddu in C. N. Reilley’s “Advances in Analytical Chemistry and Instrumentation,” Interscience Publishers, Inc., New York, N. Y . ,Vol. 1, 1960, pp. 347-424. (17) W. H. Washburn and M. J. Mahoney, J . A m . Chem. Soc., 80, 504 (1958). Recent results of Mr. S. S. Labana indicate that a variety of newly synthesized bicyclo [3.l.O]hexanes all show characteristic absorption in the 1.68 # region, attributable to cyclopropyl CH bonds.
XXIIIa (18) L. M. Jackman, “Applications of Nuclear Magnetic Resonance Spectroscopy in Organic Chemistry,” Pergamon Press. New York, N. Y.,1959, p. 52. (19) This synthesis was carried out by Professor A. W. Burgstahler and Robert Sticker, and will be described by them elsewhere. We are most grateful to the University of Kansas group for providing us with a sample of ( f ) - X X I I I .
980
JERROLD
MEINWALD, ARTHURLEWISAND PAULG. GASSMAN
With the stereochemical and rearrangement problems disposed of, we may turn to the preparation of the borneol and isoborneol analogs I11 and IV. Ozonolysis of the exocyclic olefin X, even under the conditions appropriate for the cleavage of camphene to camphenilone,20gave a mixture of a t least three products, and was therefore not studied further. Osmylation of X gave only a low yield of crude diol, and appeared unpromising. Application of the Baeyer-Villiger reaction, however, gave the desired results. The highly hindered acid I I a proved inert to methyllithium in ethyl ether and in refluxing tetrahydrofuran. However, the acid chloride derived from IIa reacted smoothly with dimethylcadmium to give the exo-ketone XXV accompanied by a small amount of the more stable endo isomer XXVI, which was synthesized
VOl. 84
depressed melting points, as well as surprisingly similar retention times. However, their distinctive specific rotations (+1.5.1' for I11 and -17.4' for IV) and their infrared spectra, which reveal striking differences in the "finger-print" region, leave no doubt that they are different substances. A further study of the chemistry of these compounds is now in progress. Acknowledgments.-The support of this research by the Alfred P. Sloan Foundation and the National Science Foundation is acknowledged with pleasure. The senior author would like to thank the John Simon Guggenheim Foundation for a Fellowship during the academic year 1960-1961, and the Department of Chemistry, Stanford University, for its cordial hospitality in the Winter and Spring 1961 Quarters. Experimental
2 , Cd(CHd2
( +)-1,5,5-Trimethylbicyclo [2.1.1]hexane-exo-6-carboxylic
A
XXV
independently as described below. Rather than use this inhomogeneous ketone for further work, the endo-acid I I b was converted into the endoketone XXVI, a reaction which proceeded readily with ethereal methyllithium. Trifluoroperacetic acid oxidation?' of XXVI followed by lithium aluminum hydride reduction gave a good yield of I11 as an optically active, crystalline solid. 1, CFaC03H 2, LiAlHd
H IIb
kOnH
VCH, XXVI O
Acid (IIa).-The photochemical ring contraction of I was carried out as described by Horner and Spietschka t o give ~ (CHC&), in about the acid 11, m.p. 100-101", [ a ]+13.8" 60% yield.3 In addition t o the acidic product, 3.16 g. of neutral material was obtained from the irradiation of 13.0 g. of I. Vapor phase chromatography (using 0.5% polyethylene glycol on glass beads a t 85", with helium as carrier) showed that a.t least two compounds (retention times, 3.00, 3.90 min.) were present in the neutral material. A volatile solid with a strong camphor-like odor was collected by sublimation, and purified by repeated sublimation, m..p. 165167"; infrared spectrum (KBr), 5.70 p. The meltlng point of isocyclenone is 168-170°.23 A semicarbazone was prepared in the usual way. Two recrystallizations from ethanol gave an analytical sample, m.p. 242.5-243" (isocyclenone semicarbazone m.p. 243244 ') .2 3 Anal. Calcd. for CllHi70N3: C, 63.74; H, 8.27; N, 20.27. Found: C,63.57; H,8.35; N, 19.90. The residue after sublimation was recrystallized five times from pentane t o give a white solid, m.p. 220'. A mixed melting point with camphoric anhydride (m.p. 220°)24 gave no depression. The infrared spectrum (KBr), 5.57, 5.70 f i , was identical with that of camphoric anhydride. ( +)-1,5,5-Trimethylbicyclo [2.1.11hexane-em-6-carboxylic Acid Methyl Ester.-An ethereal solution of diazomethane was added dropwise t o an ethereal solution of 10.00 g. of I I a until the color of the diazomethane persisted. The reaction mixture was dried over anhydrous magnesium sulfate, filtered, and the solvent removed on a flash evaporator. The residue was fractionally distilled to give 9.96 g. (92y0) of ester, b.p. 98-100" (18 mm.), nZ8D 1.45?0, [ a ] ~+29" (EtOH). This material showed a 2.52 mm. retention time on a 0.5% polyethylene glycol 400 on glass bead column (3' X 1/4") a t a temperature of 119', using helium a s carrier gas. KOimpurity was detectable. ( +)-1,5,5-Trimethylbicyclo[2.1.l]hexane-endo-6-carboxylic Acid (IIb).-To a solution of 6.00 g. of sodium hydroxide in 80 ml. of ethanol and 20 ml. of water was added 9.10 g. of the methyl ester of IIa. This solution n-as refluxed for 2 hr., diluted with 100 ml. of water and extracted thoroughly with ether. The aqueous solution was acidified with dilute hydrochloric acid and extracted with methylene chloride. The methylene chloride solution was washed with water, dried over anhydrous magnesium sulfate, filtered and the solvent removed under reduced pressure. The residue was sublimed to give 7.25 g. (867,) of IIb, m.p. 43-43.j0, [ a ] D 4-68.7" (CHCl,). Anal. Calcd. for CioHlsOz:C, 71.39; H, 9.59. Found: C, 71.61; H, 9.52. An anilide was prepared via the acid chloride. Two recrystallizations from aqueous ethanol gave an analytical sample as silky, white needles, m.p. 128-128.5'.
&FaT. \/
\ /
OH
I11
OH
XXVII
IV
H
Sarett reagentz2 converted I11 into the rather unstable cyclobutanone XXVII (A, 5.61 p), which was reduced by lithium aluminum hydride to the isoborneol analog IV. The ketones XXVI and XXVII, as well as the alcohols I11 and IV, gave good evidence of homogeneity upon gas chromatographic analysis. Both of the epimeric alcohols 111 and IV have similar and very easily ( 2 0 ) P. S . Bailey, C h e m Ber., 88, 795 (1955). (21) W.D.Emmons and G. B. Lucas, J . A m . Chem. SOL.,77, 2287 (I95,5), (22) G.I. Poos, G. E. Arth, R. E. Beyler and L. H. Sarett, ibid.,7 6 , ,422 (19,531.
(23) Elsevier's "Encyclopedia of Organic Chemistry," Elsevier Publishing Co., Inc., Xew S o r k , N. Y., S-d. 13, Sec. 111, 1946, p. 1051. (24) J. L. Simonsen, "The Terpenes." Cambridge University Press, Cambridge, England, Vol. 11, 1949. p. 480.
March 20, 1962
1,5,5-TRIMETHYLBICYCLO [2.1.1]HEXANE-6-CARBOXYLIC ACID
9Sl
Anal. Calcd. for C16HZ10N: C, 78.97; H, 8.70; N, Anal. Calcd. for Cl8HzsOsN4:C, 50.70; H, 6.15; N, 5.76. Found: C,79.13; H,8.74; N,5.88. 13.14. Found: C, 50.96; H, 6.18; N, 13.29. endo-6-Dimethylaminomethyl-1,5,5-trimethylbicyclo( +)-1,5,5-Trimethylbicyclo[Z .l.llhexane-endo-6-carboxylic acid methyl ester was prepared by treatment of 1.00 [Z.1.11hexane N-oxide was prepared by treatment of 2.15 g. of the acid I I b with diazomethane as described for the g. of the endo-amine with 3OY0 hydrogen peroxide as described for the exo epimer. exo-methyl ester to give 0.97 g. (90yo) of product, b.p. A picrate of the N-oxide was prepared in 95y0 ethanol. 97-99' (19 mm.), %"D 1.4545, [a]D +64.4' (EtOH). Two recrystallizations from 9570 ethanol gave an analytical This product showed a 2.20 min. retention time using V.P.C. under the same conditions used for the epimeric ester; sample as yellow needles, m.p. 134-135'. no impurity was detectable. Using a 6' column of the Anal. Calcd. for ClsH~sOsN4: C, 50.70; H, 6.15; ii, same type, a t 85", a retention time of 19.9 min. (cf. 21.8 13.14. Found: C, 50.81; H,6.23; N, 13.32. min. for the epimeric ester) was observed. 6-Methylene-1,5,5-trimethylbicyclo[Z.l.l]hexane(X). Anal. Calcd. for CllH1802: C, 72.49; H , 9.96. Found: A.-The N-oxide prepared from 2.70 g. of exo-amine was pyrolyzed at 130-135' under a vacuum of 50 mm. The C, 72.53; H, 9.96. ( )-1,5,5-Trimethylbicyclo [Z.1.11hexane-exo-6-di- pyrolysis products were condensed in a trap which had been The trapped methy1carboxamide.-A solution of 5.00 g. of IIa and 4.50 cooled in a Dry Iceacetone-bath to -50". g. of oxalyl chloride in 50 ml. of dry benzene was allowed material was diluted with ether, and washed with water until the aqueous solution showed no signs of basicity. to stand a t room temperature for 18 hr. The benzeneoxalyl chloride was removed under reduced pressure. The The ethereal solution was dried over anhydrous magnesium sulfate, filtered, and the ether removed by distillation. The residual acid chloride was diluted to 50 ml. with dry ether, and dimethylamine was bubbled into the solution. The di- residue was sublimed to give 1.18 g. (69'$c) of X, m.p. methylamine hydrochloride was removed by filtration, and 60-61"; infrared spectrum (neat): 3.29, 5.96, 11.56 p. the solvent was evaporated under reduced pressure to give These peaks vanish when X is allowed to react with bro5.45 g. (94%) of amide, b.p. 86-88' (1 mm.), WZ8D 1.4861, mine in carbon tetrachloride. [ a ] D $22.3" (EtOH). Anal. Calcd. for CloHla: C, 88.16; H, 11.84. Found: Anal. Calcd. for C12H210N: C, 73.79; H, 10.84; N, C, 88.37; H , 11.96. 7.17. Found: C, 73.45; H, 10.75; N, 7.28. B.-The N-oxide prepared from 2.70 g. of endo-amine ( +)-1,5,5-Trimethylbicyclo[Z .1 .l]hexane-endo-6-di- on pyrolysis under the same conditions gave 1.08 g. (63%) methylcarboxamide was prepared via the acid chloride as of X, m.p. 60-61'; infrared spectrum (neat) identical described for the exo-dimethylcarboxamide. A 4.60-g. with that of the olefin obtained from the exo isomer. Rearrangement of the 1,5,5-Trimethylbicyclo[Z.l.l]portion of the acid I I b gave 4.40 g. (82%) of product, b.p. hexane-6-carboxylic Acids.-Six grams of finely ground IIa 105-107" (3 mrn.), nZ8D1.4818, [ a ] D +19.7" (EtOH). Anal. Calcd. for C~ZHZIOX:C, 73.79; H , 10.84; AT, was added slowly to 20 ml. of vigorously stirred concentrated sulfuric acid. After the addition was complete, the re7.17. Found: C, 73.47; H, 10.80; h-,7.28. action mixture was stirred a t 45-50" for 0.5 hr., cooled, ( )-exo-6-Dimethylaminomethyl-1,5,5-trimethylbicyclo- and poured onto 250 g. of ice. The solution was neutra[Z.l.l]hexane.--A solution of 5.00 g. of em-amide in 50 ml. lized with dilute aqueous sodium hydroxide solution and exof anhydrous ether was added dropwise to a stirred suspen- tracted with ether. The ethereal solution was dried over sion of 3.50 g. of lithium aluminum hydride in 50 ml. of anhydrous magnesium sulfate, filtered, and the ether evapoanhydrous ether. The reaction mixture was stirred at room rated under reduced pressure. The residue (3.42 9.) temperature for 12 hr., cooled to 0" and quenched by was fractionally distilled to give 2.94 g. (49Yc) of XIII, dropwise addition of 14.0 g. of water. The resultant b.p. 94-96' (2 mm.). The distillate crystallized immemixture was stirred vigorously for 2 hr. and then filtered. diately to give a waxy solid, m.p. 35-37'; infrared (neat): The precipitated salts were washed thoroughly with ether, 5.64 p . Sublimation gave an analytical sample, m.p. and the combined ether extract was dried over anhydrous 36-37', [ a ] ~ $3.9" (c 1.5). The infrared spectrum of magnesium sulfate. Evaporation of the ether followed by this product was superimposable on that of an authentic fractional distillation of the residue gave 4.28 g. (90Yc) of sample of XIII.12 amine, b.p. 88-90' (18 mm.), %"D 1.4640, [ a ] ~$8.7" Anal. Calcd. for CIOHI~OZ: C, 71.43; H , 9.53; mol. (EtOH). Anal. Calcd. for C12H23S: C, 79.49; H, 12.79; N, 7.73. wt., 168. Found: C, 71.43; H , 9.44; mol. wt., 169. The basic aqueous solution was acidified with dilute Found: C,79.60; H, 13.09; N,7.23. A picrate of this amine was prepared in 95y0 ethanol. hydrochloric acid. The precipitate which formed was colThree recrystallizations from 9570 ethanol gave an analytical lected by filtration and air-dried to give 1.16 g. (19%) of XXIII; recrystallization from dilute acetic acid and subsample as yellow needles, m.p. 132-132.5'. sequent sublimation gave an analytical sample, m.p. Anal. Calcd. for C18H2s07N,: C, 52.67; H , 6.39; N, 80-81", [ a ] ~-26.9" (c 1.3) (chloroform); ultraviolet 13.65. Found: C,52.79; H,6.43; N, 13.86. spectrum: a t 210 my E = 600. This acid gave no color with ( +)-endo-6-Dimethylamiomethyl-1,5,5-trimethylbicyclo- tetranitromethane, and failed to react with bromine, [Z.Z.l]hexane.-The endo-amine was prepared by lithium potassium permanganate and ozone. aluminum hvdride reduction as described for the exoAnal. Calcd. for C1OH1602: C, 71.43; H, 9.53. Found: dimethylamine. Reduction of 3.60 g. of the endo-amide C, 71.32; H, 9.63. gave 3.16 g. (94%) of amine, b.p. 86-87' (19 mm.), n% The infrared spectrum of this material was identical to 1.4588, [ a ] D $27.6" (EtOH). that of the synthetic sample of &-XXIII provided by Anal. Calcd. for C12Hi3N: C, 79.49; H , 12.79; N, 7.73. Burgstahler and Sticker.lg Found: C,79.67; H, 12.82; N,8.06. The same two products mere formed in comparable -4 picrate was prepared in 95% ethanol. Three recrystal- amounts when IIb was used in place of IIa. lizations from 95y0 ethanol gave an analytical sample as Lithium Aluminum Hydride Reduction of XIII.-.4 yellow needles, m.p. 126-127'. solution of 0.30 g. of XI11 in 15 ml. of anhydrous ether was Anal. Calcd. for C I S H ~ ~ O ~C,X ~52.67; : H, 6.39; N, added dropwise to a stirred suspension of 0.30 g. of lithium aluminum hydride in 5 ml. of anhydrous ether. Stirring 13.65. Found: C, 52.69; H , 6.46; N, 13.51. continued fur 6 hr. a t room temperature. n'ater exo-6-Dimethylaminomethyl-1,5,5-trimethylbicyclo- was [Z.l.l]hexane N-oxide.-To a solution of 3.05 g. of exo- (1.20 8.) wa5 added dropwise and the resulting inorganic amine in 7 ml. of methyl alcohol was added, with stirring salts were removed by filtration and washed thoroughly and cooling, 8 ml. of 30% hydrogen peroxide. After stir- with ether. The solution was dried over anhydrous magnesium sulfate, filtered and the solvent was removed to yield ring for an additional 12 hr., a small amount of platinum 0.22 g. (72Yc) of XIV. Recrystallization from 6S70' black was added, and the mixture was stirred for 24 hr. to destroy the excess hydrogen peroxide. The solution was petroleum ether gave a sample of the diol, m.p. 145-145.5", then filtered and the solvent evaporated in a stream of dry m.m.p. (with authentic sample of XIV12) 145-145.5'. Anal. Calcd. for CloHpaOt: C, 69.72; H , 11.70. Found: air. C, 69.90; H, 11.88. A picrate of the residual h--oxide was prepared in 957, ethanol. Two recrystallizations from 9570 ethanol gave 1,2,2-Trimethylbicycl0[3.1 .O] hexane-6-carboxylic Acid an analytical sample as yellow needles, m.p. 136-137'. Methyl Ester.-An ethereal solution containing 45 mg. of
+
+
982
JERROLD MEINWALD, ARTHUR LEWISAND PAULG. GASSMAN
Vol. s4
XXIII was treated with an excess of ethereal diazomethane. Osmylation of 6-Methylene-l,5,5-trimethylbicyclo [ 2.1.11 The solvent was allowed to evaporate and an analytical hexane.-A solution of 1.00 g. of osmium tetroxide in 10 sample, n a a D 1.4608, was prepared by gas chromatography. ml. of dry dioxane was added to a solution of 0.50 g. of X Infrared analysis showed no olefinic carbon-hydrogen ab- in 10 ml. of dry dioxane. The stoppered reaction mixture sorption, and the spectrum was identical to that of the ester was kept at room temperature for 4 days in the dark. The reaction mixture was saturated with hydrogen sulfide, prepared from Burgstahler and Sticker's (&)-XXIII. Anal. Calcd. for CllHlsOz: C, 72.35; H, 9.89. Found: filtered, and the precipitated salts washed with dioxane. The combined filtrates were taken to dryness under reduced C, 72.24; H , 10.17. to give 0.20 g. of a semi-solid oil; infrared spectrum Dihydro-6-campholenolacetonefrom XXII1.-Sixty milli- pressure 3.00,5.70-5.80p. grams of XXIII and 2 ml. of concentrated sulfuric acid were (neat): exo-6-Acetyl-l,5,5-trimethylbicycl0[2.1.1] hexane (XXV). carefully mixed and stirred a t 45-50" for 45 min. The A.-The acid 1Ia was inert to prolonged treatment with reaction mixture was poured onto ice, made basic with dilute potassium hydroxide solution and extracted with ether. excess methyllithium in both refluxing ether and refluxing The ethereal solution was dried and then ether distilled tetrahydrofuran. B.-To a solution of methylmagnesium iodide prepared off to give 30 mg. of XIII. in the usual way from 7.60 g. of methyl iodide and 1.28 g. 1,2,2-Trimethyl-6-acetylbicyclo[3.1 .O] hexane.-The magnesium metal in 100 ml. of anhydrous ether was added methyllithium prepared from 0.70 g. of methyl iodide, of 10.0 g. of anhydrous cadmium chloride during 10 min. The 0.10 g. of lithium and 15 ml. of anhydrous ether was added mixture was stirred and heated under reflux for 45 min. dropwise to a stirred solution of 300 mg. of XXIII in 10 ml. The ether was distilled from the reaction mixture and reof anhydrous ether. After stirring for 1 hr., the reaction placed by 100 ml. of dry benzene. About 10 mg. of anhymixture was diluted with water and ether, and the layers drous chloride26 was added, and the reaction mixture were separated. Acidification of the basic aqueous layer cooled ferric in an ice-bath. The acid chloride prepared from 8.00 gave 166 mg. of unreacted starting acid XXIII. The g. of 1Ia and oxalyl chloride, as described above, was added ethereal solution was dried over anhydrous magnesium sulfate, filtered, and the ether removed under reduced pres- dropwise. After the addition was complete the mixture was stirred and refluxed for 2 hr. The reaction mixture sure to give 106 mg. (80% based on unrecovered XXIII) of cooled in an ice-bath and 100 ml. of water was added the desired methyl ketone; infrared spectrum (neat): was cautiously. The aqueous phase was separated and extracted 5.90 p ; ultraviolet spectrum (ethanol): AXmaI 206 mp with ether. The combined organic layers were washed with (e5,700), 278mp((e67). water, dried over anhydrous magnesium sulfate, filtered A 2,4-dinitrophenylhydrazonewas prepared in the usual and the solvent removed under reduced pressure. The way. Four recrystallizations from ethanol gave bright residue was fractionally distilled to give 6.24 g. (78%) of ultraviolet spectrum an oil, b.p. orange needles, m.p. 156-157'; 85-87' (11 mm.); infrared spectrum (neat): AAmax235mp ( e 11,800), 370mp(e 17,600). (ethanol): 5.88 p , with no absorption in the hydroxyl region of the Anal. Calcd. for C17H22N4O4: C, 58.94; H, 6.40; N, spectrum. 16.18. Found: C, 58.98; H, 6.53; N, 16.32. Anal. Calcd. for CllHleO: C, 79.46; H, 10.92. Found: l-Methylbicyclo[3. 1.O]hexane-6-carboxylicAcid (XXIV). C, 79.65; H, 10.93. -A mixture of 4.10 g. of l-methylcyclopentene,z6 2.85 g. A semicarbazone was prepared in the usual way. Three sf ethyl diazoacetate and 0.28 g. of powdered copper was recrystallizations heated on a steam-bath for about 4 min., a t which time a m. p .209.52 10.5".from ethanol gave an analytical sample, vigorous exothermic reaction started. The reaction mixAnal. Calcd. for C12Hz10N3: C, 64.51; H, 9.48; N, ture was cooled to room temperature where the rate of 18.82. Found: C,64.52; H,9.47; K, 19.02. nitrogen evolution continued to be extremely rapid. When A 2,4-dinitrophenylhydrazonecould not be formed in the the gas evolution ceased, the reaction mixture was immersed in an ice-bath and an additional 2.85 g. of ethyl normal way. The methyl ketone was heated with the readiazoacetate was added. Vigorous evolution of gas occurred gent in ethanol a t 70' for several minutes. On cooling, for about 30 min. The reaction mixture was filtered and a derivative was obtained. Two recrystallizations from fractionally distilled to give 3.07 g. (36%) of crude ester, ethanol gave yellow crystals, m.p. 131-132O. The melting b.p. 94-101' (15 mm.). This material could not be ob- point was not depressed on mixing with the 2,4-dinitrotained analytically pure, but was hydrolyzed directly to the phenylhydrazone obtained in the normal way from the epimeric endo-methyl ketone XXVI. acid. A mixture of 1 g. of the distilled ester, 2 g. of potassium Although the ketonic product was analytically pure, vapor hydroxide, 10 ml. of water and 10 ml. of methanol was re- phase chromatography showed it to be a mixture of 80% fluxed on a steam-bath for 3 hr. The methanol was re- of the desired methyl ketone XXV and 207' of the epimeric moved on a flash evaporator, and the residual aqueous endo-methyl ketone XXVI (retention times 12.5 and 10.3 solution was acidified with dilute hydrochloric acid. The min., respectively, on a 6' X 0.25" column a t loo", packed precipitate which formed was collected by filtration and air- with 0.37, polyethylene glycol 400 on glass beads, using helium as carrier). dried to give 0.27 g. (32%) of crude XXIV, m.p. 50-53'. (f)-endo-6-Acetyl-1,5,5-trimethylbicyclo[2.1 .l]hexane Recrystallization from nitromethane, followed by subli(XXVI).-Methyllithium was prepared by adding a solumation, gave an analytical sample, m.p. 62.5-63.0'. Anal. Calcd. for COH1t02:C, 68.54; H, 8.63. Found: tion of 6.80 E. of methvl iodide in 35 ml. of anhydrous ether to a stirred &spension-of 0.70 g. of lithium ribbon in 35 ml. C, 68.56; H , 8.66. of anhydrous ether a t a rate necessary to maintain mild Ozonolysis of 6-Methylene-1,5,5-trimethylbicyclo[2.1.1]- refluxing. hexane.-A solution of 1.18 g. of X in 10 ml. of absolute The resultant solution of methyllithium was added dropmethanol was exhaustively ozonized a t -70 to -80°.20 wise to a stirred solution of 3.36 g. of IIb in 50 ml. of anhyThe ozonized solution was then added to a vigorously drous ether, during 1 hr. After stirring for an additional stirred mixture of 7.5 ml. of absolute methanol, 2.5 ml. of the reaction mixture was poured onto 75 g. of crushed glacial acetic acid and 4.0 g. of sodium iodide, and stirred hour, ice and extracted wlth ether. The ethereal solution was overnight. The mixture was diluted with 25 ml. of water, dried over anhydrous magnesium sulfate, filtered and the and sufficient sodium bicarbonate was added to make it solvent removed under reduced pressure. The residue was basic. The reaction mixture was then extracted with ether, fractionally distilled to give 2.75 g. (827') of XXVI, b.p. and the combined ethereal extracts dried over anhydrous 80-82' (12 mm.), ~ Z S D 1.4618, [a]D $34.6" ,(EtOH); magnesium sulfate. The drying agent was removed by spectrum (neat): 5.88 p , with no absorptlon in the filtration and the solvent evaporated under reduced pressure infrared hydroxyl region of the spectrum. Vapor phase chromato give 0.86 g. of an oil; infrared spectrum (neat): 2.90, tography, carried out as described for XXV, showed a 5.80 p . single peak, 10.3 min. retention time. Vapor phase chromatography under the same conditions Anal. Calcd. for CllHlsO: C, 79.46; H, 10.92. Found: described for the two methyl esters showed the product to be a mixture of a t least three components, retention times C, 79.56; H, 11.00. 4.30, 7.58 and 10.28 min. (25) G. Chavanne and L. de Vogel, Bull. soc. chim. Belg., 37, 142
(1928).
(26) T h e authors are indebted to Professor James Cason for calling their attention to the catalytic effect of ferric chloride on the reaction of dimethylcadmium w i t h acid chlorides.
March 20, 1962
DEHYDROBROMINATION OF BIPHENYLYL BROMOCYCLOHEXYL KETONE
A semicarbazone was formed in the usual way. Two recrystallizations from ethanol gave an analytical sample, m.p. 221-222'. Anal. Calcd. for C1?H210Na: C, 64.54; H, 9.48; N, 18.82. Found: C, 64.46; H,9.45; N, 18.96. A 2,4-dinitrophenylhydrazone was formed in the normal way. Three recrystallizations from ethanol gave an analytical sample, m.p. 131-132'. A n d . Calcd. for C17H~204N,: C, 58.94; H , 6.40; N, 16.18. Found: C, 58.89; H,6.25; N, 15.89. ( )-1,5,5-Trimethylbicyclo[2.1.1] hexan-endo-6-01(III).A solution of trifluoroperacetic acid*' was prepared by mixing 3.40 ml. of trifluoroacetic anhydride with 0.52 ml. of 90% hydrogen peroxide in 6.60 ml. of methylene chloride. The resulting peracid solution was added dropwise to a stirred solution of 1.60 g. of XXVI, 3.90 g. of anhydrous dibasic sodium phosphate and 20 ml. of methylene chlaride. The reaction mixture was stirred and refluxed for 8 hr., filtered, and the inorganic salts washed thoroughly with methylene chloride. The filtrate was dried over anhydrous magnesium sulfate, filtered, and the solvent removed under reduced pressure. The infrared spectrum of the residue indicated some unchanged ketone was still present. The total reaction product was, therefore, recycled by dissolving it in 20 ml. of methylene chloride to which 3.90 g. of anhydrous dibasic sodium phosphate had been added. The trifiuoroperacetic acid obtained from 3.40 ml. of tritluoroacetic anhydride, 0.52 ml. of 90% hydrogen peroxide and 6.60 ml. of methylene chloride was added dropwise. Refluxing was maintained for 12 hr. The solution was cooled, filtered and the inorganic salts washed thoroughly with methylene chloride. The filtrate was dried over anhydrous magnesium sulfate, filtered, and the solvent removed under reduced pressure to give 1.83 g. of acetate (containing some trifluoroacetate, as indicated by infrared absorption a t 5.60 as well as 5 . 7 8 ~ ) . The crude acetate was dissolved in 10 ml. of anhydrous ether and added dropwise to a stirred suspension of 1.00 g. of lithium aluminum hydride in 20 ml. of anhydrous ether. After stirring a t room temperature for 8 hr., the reaction mixture was cooled to 0" and quenched by the dropwise addition of 4.0 ml. of water. The hydrolyzed mixture was stirred for 1 hr. and the precipitated salts were removed by filtration. The salts were washed thoroughly with ether, and the combined filtrates were dried over anhydrous magnesium sulfate. The drying agent was removed by filtration and the solvent was removed under reduced pressure. Sublimation of the crude product gave 1.01 g.
+
983
(73% yield based on XXVI) of 111, m.p. 95-97' with preliminary softening a t 85', [a]D 15.1' (EtOH). Preparative vapor phase chromatography gave an analytically pure sample, m.p. 121-122'. This material showed a 2.45 min. retention time on a column of the type used for the methyl esters, operated a t 100'. No impurity was detectable. Anal. Calcd. for C9H160: C, 77.09; H , 11.50. Found: C, 77.00; H, 11.39. A p-nitrobenzoate was formed in the normal way. Two recrystallizations from aqueous ethanol gave an analytical sample, m.p. 98.599'. Anal. Calcd. for ClsHlgOdN: C, 66.42; H , 6.62; N, 4.84. Found: C, 66.65; H , 6.63; N, 5.06. 1,5,5-Trimethylbicyclo[Z.1.11hexan-6-one (=VI).To the complex formed from 2.70 g. of chromium trioxide in 40 ml. of dry pyridine**was added dropwise with stirring a solution of 1.50 g. of I11 in 20 ml. of dry pyridine. The resulting mixture was stirred a t room temperature for 3 days, and then 160 ml. of water was added. The mixture was extracted with ether, and the ether washed repeatedly with water and then dried over anhydrous magnesium sulfate. The drying agent was removed by filtration and the solvent was evaporated under reduced pressure to give 0.76 g. (51y0) of XXVII; infrared spectrum (neat): 5.61 p , with no absorption in the hydroxyl region of the spectrum. This ketone showed a surprising tendency to decompose upon distillation. ( - )-1,5,5-Trimetnylbicyclo[2.1.1]hexan-exo-6-01 (IV).A solution of 0.75 II. of crude XXVII in 15 ml. of anhvdrous ether was added aropwise to a stirred suspension of 0.50 g. of lithium aluminum hydride in 25 ml. of anhydrous After stirring for 2 hr. a t -5', the mixture ether a t -5". was hydrolyzed by the dropwise addition of 2.0 g. of water. The reaction mixture was filtered and the inorganic salts washed thoroughly with ether. The combined filtrates were dried over anhydrous magnesium sulfate, filtered and the solvent removed under reduced pressure. The residue was sublimed under reduced pressure to give 0.56 g. (73%) of a waxy white solid. Preparative vapor phase chromatography gave an analytical sample, m.p. 124-125.5', [m]D -17.4'. This material showed a V.P.C. retention time of 2.55 min., with no detectable impurity, under the conditions used to examine its epimer 111. Anal. Calcd. for CgH160: C, 77.09; H, 11.50. Found: C, 77.15; H, 11.62. An attempt to form a p-nitrobenzoate in the normal way was unsuccessful.
+
[CONTRIBUTIONFROM AVERYLABORATORY, UNIVERSITY
OF
NEBRASKA, LINCOLN, hTEBR.1
Elimination Reactions of a-Halogenated Ketones. VI1.I Kinetics of Several Dehydrobromination Reactions of 4-Biphenylyl 1-Bromocyclohexyl Ketone BY DENNISN. KEVILL,PATRICK H. HESS, PETERW. FOSTER AND NORMAN H. CROMWELL RECEIVED DECEMBER 9, 1961 The a-bromoketone 4-biphenylyl 1-bromocyclohexyl ketone ( I ) has been found to undergo elimination reaction in solvent acetonitrile and in solvent benzene to yield the a,B-unsaturated ketone 4-biphenylyl 1-cyclohexenyl ketone (11). The kinetics of the elimination reaction as promoted by piperidine, by morpholine, by tetraethylammonium bromide and by piperidine hydrobromide are reported. The kinetic results are discussed in terms of possible reaction mechanism. 4-Biphenylyl 2-piperidinocyclohexyl ketone (111) and 4-biphenylyl 2-bromocyclohexyl ketone (IV) have been prepared and characterized.
Introduction.-This communication reports upon the kinetics of several elimination reactions of 4Iiphenylyl 1-bromocyclohexyl ketone (I) in solvent acetonitrile and in solvent benzene. Elimination from 4-biphenylyl 1-bromocyclohexyl ketone (I) produces the a$-unsaturated ketone 4-biphenylyl 1-cyclohexenyl ketone (11). 2 Previous communications in this series have re(1) For paper V1 in this series see D. N. Kevill and N. H. Cromwell, J , A m . Chem. Soc., 83, 3815 (1961). (21 X. H. Cromwell and P. H. Hew. ibid., 82, 136 (1960).
ported upon the kinetics of elimination reactions of 2-bromo-2-benzyl-4,4-dimethyl-l-tetralone (V) in solvent acetonitrile.a91A merged substitution and elimination mechanism was proposed for these elimination reactions. Discussion of Results The decompositions of a-bromoketone I, of piperidine and of morpholine in both acetonitrile and in benzene were found, even a t 90.6', to be (3) D. N. Kevill and N. H.Cromwell, ibid., 83, 3812 (1961).
